Mount Sinai Researchers Enhance AI Algorithm to Improve Detection and Risk Assessment of Hypertrophic Cardiomyopathy

Mount Sinai Researchers Enhance AI Algorithm for Heart Disease Detection

Researchers at Mount Sinai have made significant strides in improving the detection and risk assessment of hypertrophic cardiomyopathy (HCM), a type of heart disease affecting one in 200 people worldwide. The team has calibrated an artificial intelligence (AI) algorithm, known as Viz HCM, to provide more precise and actionable information for both clinicians and patients 1

1

Improved Risk Assessment and Patient Communication

The enhanced AI tool now assigns numeric probabilities to its findings, allowing for more specific risk assessments. For instance, instead of simply flagging a patient as "high risk for HCM," the algorithm can now provide a more nuanced assessment, such as "You have about a 60 percent chance of having HCM" 2

2

Dr. Joshua Lampert, Director of Machine Learning at Mount Sinai Fuster Heart Hospital, emphasized the importance of this development: "This is an important step forward in translating novel deep-learning algorithms into clinical practice by providing clinicians and patients with more meaningful information" 3

3

Study Methodology and Results

The research team applied the Viz HCM algorithm to nearly 71,000 patient electrocardiograms (ECGs) recorded between March 2023 and January 2024. The algorithm identified 1,522 patients with a positive alert for HCM. Researchers then reviewed medical records and imaging data to confirm HCM diagnoses 1

1

After confirming the diagnoses, the team applied model calibration to the AI tool. They found that the calibrated model provided an accurate estimate of a patient's likelihood of having HCM, correlating well with actual disease prevalence 2

2

Related Stories

AI Revolutionizes Heart Disease Detection: From ECGs to Hidden Conditions

AI-Enhanced ECG and CT Scans Revolutionize Cardiovascular Risk Prediction

AI Model Outperforms Doctors in Predicting Sudden Cardiac Death Risk

AI Model Detects Multiple Heart Failure Types from Routine ECGs, Including Hard-to-Identify HFpEF

Clinical Implications and Future Directions

The enhanced AI tool has the potential to transform clinical practice by allowing cardiologists to prioritize high-risk patients for earlier appointments and treatment. This proactive approach could help prevent complications associated with HCM, such as sudden cardiac death or symptoms from thickened heart muscle obstructing blood flow 3

3

Dr. Vivek Reddy, Director of Cardiac Arrhythmia Services for the Mount Sinai Health System, commented on the study's implications: "This study provides much-needed granularity to help rethink how we triage, risk-stratify, and counsel patients. In an era of augmented intelligence, we must grow to incorporate novel sophistication in our approach to patient care" 1

1

The research team plans to expand this study and AI calibration for HCM to additional health systems across the country, further validating and refining the tool's effectiveness in diverse clinical settings 2

2

Ethical Considerations and Disclosure

It's worth noting that Viz.ai sponsored this study, and Dr. Lampert is a paid consultant for the company. This relationship highlights the importance of transparency in AI research and the need for careful consideration of potential conflicts of interest in medical AI development 4

4